Deep Learning for Vision – Foundations and Applications

This course provides a comprehensive introduction to the key concepts and architectures underlying modern deep learning for image data. It covers the mathematical and computational foundations of  vision architectures such as convolutional networks, vision transformers, and hybrid architectures. Students study the principles behind image classification, segmentation, detection, and tracking, along with low-level tasks like denoising and super-resolution. The course further will give an introduction into advanced topics such as self-supervised representation learning, generative modeling, and multimodal vision, emphasizing conceptual understanding and the unifying ideas that drive state-of-the-art vision research.

Modules:

• CMS-AAI-AP
• CMS-CLS-ELV
• CMS-VC-ELV1

Location: BAR/0E85/U
Time: 
       Lecture:  Fri 4 DS. 1pm - 2:30pm BAR/0E85/U
       Exercise: Thu 5 DS 2:50pm - 4:20pm S14/745

First Lecture/Coordination meeting: 
Fri, Oct. 17th 2025 4DS 1pm-2:30pm BAR/0E85/U

First exercise Session:
Thu, Oct 23rd 2025, 5DS online